Classification of Land-use Based on Remote Sensing Image Texture Features with Multi-scales and Cardinal Direction Inspired by Domain Knowledge

doi:10.11947/j.AGCS.2016.20150624

Abstract

Abstract: Texture features based on grey level co-occurrence matrix (GLCM) are effective for image analysis, and this paper proposed a new method to construct GLCM with multi-scales and cardinal direction factors inspired by domain knowledge, in order to improve the performance of texture features and solve the uncertainty problems in image classification of land-use. By simulating the process of human visual interpretation, an integrated computation pattern of GIS and RS data were performed. Firstly, on the basis of image registration, some classic GIS spatial data mining algorithms were employed to asymptotically extract domain morphological knowledge; Next, under the responding mechanism derived from correlated analysis, an algorithm for establishing GLCM multi-scale windows that can match categories one by one, an algorithm for determining GLCM weighted cardinal direction windows that can describe observation orientation were designed based on relevant morphology indexes. Experimental results indicate that, there is a strong correlation between domain morphological knowledge and GLCM construction factors, meanwhile, with lower computational complexity, the new method can extract stable texture features to describe actual spatial meanings of complex objects, thereby improve the image classification accuracy of land-use.

Key words: GLCM texture image classification, multi-scale windows, weighted cardinal direction, integrated computing, GIS spatial data mining

CLC Number:

P208

LAN Zeying, LIU Yang. Classification of Land-use Based on Remote Sensing Image Texture Features with Multi-scales and Cardinal Direction Inspired by Domain Knowledge[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(8): 973-982.

References

[1] HARALICK R M,SHANMUGAM K,DINSTEIN I H.Textural Features for Image Classification[J].IEEE Transactions on Systems,Man,and Cybernetics,1973,SMC-3(6):610-621.
[2] GONG P,HOWARTH P J.Frequency-based Contextual Classification and Gray-level Vector Reduction for Land-use Identification[J].Photogrammetric Engineering and Remote Sensing,1992,58(4):423-437.
[3] 崔林丽.遥感影像解译特征的综合分析与评价[D].北京:中国科学院遥感应用研究所,2005.CUI Linli.Integrative Analysis and Evaluation of the Interpretation Features in Remote Sensing Image[D].Beijing:Institute of Remote Sensing Applications,Chinese Academy of Sciences,2005.
[4] 陈忠.高分辨率遥感图像分类技术研究[D].北京:中国科学院遥感应用研究所,2006.CHEN Zhong.Research on High Resolution Remote Sensing Image Classification Technology[D].Beijing:Institute of Remote Sensing Applications,Chinese Academy of Sciences,2006.
[5] FRANKLIN S E,WULDER M A,LAVIGNE M B.Automated Derivation of Geographic Window Sizes for Use in Remote Sensing Digital Image Texture Analysis[J].Computers&Geosciences,1996,22(6):665-673.
[6] 马洪超,李德仁.基于空间统计学的空间数据窗口大小的确定[J].武汉大学学报(信息科学版),2001,26(1):18-23.MA Hongchao,LI Deren.Geographic Window Fixing and Its Application Based on Spatial Statistics[J].Geomatics and Information Science of Wuhan University,2001,26(1):18-23.
[7] CHICA-OLMO M,ABARCA-HERNáNDEZ F.Computing Geostatistical Image Texture for Remotely Sensed Data Classification[J].Computers&Geosciences,2000,26(4):373-383.
[8] GOODCHILD M F.Metrics of Scale in Remote Sensing and GIS[J].International Journal of Applied Earth Observation and Geoinformation,2001,3(2):114-120.
[9] PENG Bo,ZHANG Lei,ZHANG D.A Survey of Graph Theoretical Approaches to Image Segmentation[J].Pattern Recognition,2013,46(3):1020-1038.
[10] 巫兆聪,胡忠文,张谦,等.结合光谱纹理与形状结构信息的遥感影像分割方法[J].测绘学报,2014,42(1):44-50.WU Zhaocong,HU Zhongwen,ZHANG Qian,et al.On Combining Spectral,Textural and Shape Features for Remote Sensing Image Segmentation[J].Acta Geodaetica et Cartographica Sinica,2014,42(1):44-50.
[11] 骆剑承,梁怡,周成虎.基于尺度空间的分层聚类方法及其在遥感影像分类中的应用[J].测绘学报,1999,28(4):319-324.LUO Jiancheng,LEUNG Yee,ZHOU Chenghu.Scale Space Based Hierarchical Clustering Method and Its Application to Remotely Sensed Data Classification[J].Acta Geodaetica et Cartographica Sinica,1999,28(4):319-324.
[12] 张桂峰.粒度理论下的多尺度遥感影像分割[D].武汉:武汉大学,2010.ZHANG Guifeng.Multi-scale Remote Sensing Image Segmentation Based on Granularity Theory[D].Wuhan:Wuhan University,2010.
[13] 金飞,张占睦,芮杰.纹理主方向的遥感影像居民地提取[J].测绘科学,2010,35(4):139-141.JIN Fei,ZHANG Zhanmu,RUI Jie.Residential Area Extraction from Remote Sensing Image Based on Texture Principal Directions[J].Science of Surveying and Mapping,2010,35(4):139-141.
[14] 刘天时,肖敏敏,李湘眷.融合方向测度和灰度共生矩阵的纹理特征提取算法研究[J].科学技术与工程,2014,14(32):271-275.LIU Tianshi,XIAO Minmin,LI Xiangjuan.Textural Feature Extraction Algorithm Fused with Direction Measure and GLCM Method[J].Science Technology and Engineering,2014,14(32):271-275.
[15] 于晓晗,袁保宗.方向测度及其在纹理识别中的应用[J].自动化学报,1990,16(4):347-351.YU Xiaohan,YUAN Baozong.Direction Measure and Its Application to Texture Classification[J].Acta Automatica Sinica,1990,16(4):347-351.
[16] 雷小奇,王卫星,赖均.一种基于形状特征进行高分辨率遥感影像道路提取方法[J].测绘学报,2009,38(5):457-465.LEI Xiaoqi,WANG Weixing,LAI Jun.A Method of Road Extraction from High-resolution Remote Sensing Images Based on Shape Features[J].Acta Geodaetica et Cartographica Sinica,2009,38(5):457-465.
[17] 蔡利平,史文中,张华,等.边缘走向自适应的多尺度分水岭分割算法[J].中国矿业大学学报,2015,44(4):739-746.CAI Liping,SHI Wenzhong,ZHANG Hua,et al.An Edge Direction Adaptive Multi-scale Watershed Segmentation Algorithm[J].Journal of China University of Mining&Technology,2015,44(4):739-746.
[18] 李德仁,王树良,李德毅.空间数据挖掘理论与应用[M].北京:科学出版社,2006.LI Deren,WANG Shuliang,LI Deyi.Spatial Data Mining Theories and Applications[M].Beijing:Science Press,2006.
[19] 杨存健,周成虎.基于知识的遥感图像分类方法的探讨[J].地理学与国土研究,2001,17(1):72-77.YANG Cunjian,ZHOU Chenghu.Investigation on Classification of Remote Sensing Image on Basis of Knowledge[J].Geography and Territorial Research,2001,17(1):72-77.
[20] 术洪磊.GIS辅助下的基于知识的遥感影像分类方法研究--以土地覆盖/土地利用类型为例[J].测绘学报,1997,26(4):328-336.SHU Honglei.Knowledge Based Image-classification Approach Supported by a GIS[J].Acta Geodaetica et Cartographica Sinica,1997,26(4):328-336.
[21] 吴田军,骆剑承,夏列钢,等.迁移学习支持下的遥感影像对象级分类样本自动选择方法[J].测绘学报,2014,43(9):908-916.DOI:10.13485/j.cnki.11-2089.2014.0163.WU Tianjun,LUO Jiancheng,XIA Liegang,et al.An Automatic Sample Collection Method for Object-oriented Classification of Remotely Sensed Imageries Based on Transfer Learning[J].Acta Geodaetica et Cartographica Sinica,2014,43(9):908-916.DOI:10.13485/j.cnki.11-2089.2014.0163.
[22] 韩敏,孙杨.一种基于分组式蛇模型的GIS矢量边界更新方法[J].测绘学报,2009,38(2):168-174.HAN Min,SUN Yang.A Method of Vector Edge Updating Based on Grouping Snake Model in GIS[J].Acta Geodaetica et Cartographica Sinica,2009,38(2):168-174.
[23] 朱长明,骆剑承,沈占锋,等.DEM辅助下的河道细小线性水体自适应迭代提取[J].测绘学报,2013,42(2):277-283.ZHU Changming,LUO Jiancheng,SHEN Zhanfeng,et al.River Linear Water Adaptive Auto-extraction on Remote Sensing Image Aided by DEM[J].Acta Geodaetica et Cartographica Sinica,2013,42(2):277-283.
[24] 张晓东.基于遥感影像与GIS数据的变化检测理论和方法研究[D].武汉:武汉大学,2005.ZHANG Xiaodong.The Theory and Methods of Change Detection Based on Remotely Sensed Imagery and GIS[D].Wuhan:Wuhan University,2005.
[25] 张晓东,李德仁,龚健雅,等.一种基于面特征的遥感影像与GIS数据配准方法[J].遥感学报,2006,10(3):373-380.ZHANG Xiaodong,LI Deren,GONG Jianya,et al.A Matching Method of Remote Sensing Image and GIS Data Based on Area Feature[J].Journal of Remote Sensing,2006,10(3):373-380.
[26] 朱晓芳.基于支持向量机的田间杂草识别方法研究[D].镇江:江苏大学,2010.ZHU Xiaofang.Research of Weed Recognition Method Based on Support Vector Machine[D].Zhenjiang:Jiangsu University,2010.
[27] ULABY F T,KOUYATE F,BRISCO B,et al.Textural Information in SAR Images[J].IEEE Transactions on Geoscience and Remote Sensing,1986,24(2):235-245.
[28] 尹洪涛,付平,沙学军.基于DCT和线性判别分析的人脸识别[J].电子学报,2009,37(10):2211-2214.YIN Hongtao,FU Ping,SHA Xuejun.Face Recognition Based on DCT and LDA[J].Acta Electronica Sinica,2009,37(10):2211-2214.
[29] 邬建国.景观生态学--概念与理论[J].生态学杂志,2000,19(1):42-52.WU Jianguo.Landscape Ecology:Concepts and Theories[J].Chinese Journal of Ecology,2000,19(1):42-52.